P1: VEN

June 29, 2000 22:0 Annual Reviews AR103-02

Annu. Rev. Nutr. 2000. 20:2144

Copyright
c 2000 by Annual Reviews. All rights reserved

THE BEHAVIORAL DETERMINANTS OF EXERCISE:

Implications for Physical Activity Interventions

Nanc y E . She rwood and Robert W. J effe r y

Universit y of Minnesota , School of Publi c Health , Division of Epidemiolog y, Minneapolis,
Minnesot a 55454-1015; e-mail : sherwood@epi.umn.edu , jeffery@epi.umn.edu

Key Words personal characteristics, barriers, self-efficacy, motivation,

social support
Abstract In light of the well-documented health benefits of physical activity
and the fact that the majority of adult men and women are inactive, promoting regu-
lar physical activity is a public health priority. This chapter reviews current research
findings regarding the determinants of exercise behavior. It also discusses the im-
plications of this knowledge for individual and public health recommendations and
intervention strategies for promoting physical activity. The discussion is predicated
on the belief that physical activity is a complex, dynamic process. During their lives,
individuals typically move through various phases of exercise participation that are
determined by diverse factors. This chapter discusses physical activity determinants in
two broad categories: individual characteristics, including motivations, self-efficacy,
exercise history, skills, and other health behaviors; and environmental characteristics
such as access, cost, and time barriers and social and cultural supports.

CONTENTS
INTRODUCTION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
CHAPTER PURPOSE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
DIFFERENTIATING BETWEEN EXERCISE ADOPTION,
MAINTENANCE, AND RELAPSE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
PERSONAL CHARACTERISTICS ASSOCIATED WITH
PHYSICAL ACTIVITY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24
Motivation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24
Self-Efficacy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25
Stage of Change . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25
Exercise History . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25
Body Weight . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26
Health Risk Profiles . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26
Diet . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27
Stress . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27
ENVIRONMENTAL CHARACTERISTICS ASSOCIATED WITH
PHYSICAL ACTIVITY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28

0199-9885/00/0715-0021$14.00 21
P1: VEN
June 29, 2000 22:0 Annual Reviews AR103-02

22 SHERWOOD JEFFERY

Social Support . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28
Time . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29
Access . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29
Attributes of Exercise Behavior . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30
Injury . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31
INTERACTION OF PHYSICAL ACTIVITY DETERMINANTS . . . . . . . . . . . . . . 31
Summary and Implications for Physical Activity Promotion
Recommendations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32
Motivation for Physical Activity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33
Self-Efficacy for Physical Activity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34
Readiness for Physical Activity Change . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34
Barriers to Physical Activity: Time and Access . . . . . . . . . . . . . . . . . . . . . . . . . . 35
Enhancing Social Support for Physical Activity . . . . . . . . . . . . . . . . . . . . . . . . . 36
Environmental Strategies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36
SUMMARY AND CONCLUSION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37

INTRODUCTION
A physically active lifestyle has many benefits, including reduced risk of coro-
nary heart disease, hypertension, colorectal cancer, obesity, and osteoporosis
(10, 11, 14, 70). Benefits also include reduced stress and depression and increased
emotional well-being, energy level, self-confidence, and satisfaction with social
activity (43). Benefits of physical activity are evident at low- as well as high-
intensity activity levels (19, 39, 69, 90, 91, 99). Indeed, some have argued that the
greatest potential for health benefits would accrue by having sedentary adults be-
come moderately active (17, 93, 97). Physical activity appears to confer substantial
benefits at any age and regardless of prior physical activity history.
Despite the well-documented health benefits of physical activity, current esti-
mates suggest that we are in the midst of an epidemic of sedentary behavior. Mech-
anization of work and of many domestic chores has, for the most part, eliminated
obligatory physical activity from modern life. Voluntary physical activity, also
called recreational physical activity, has thus assumed central importance in filling
physical activity needs and is the focus of this chapter. Unfortunately, however,
voluntary physical activity is not popular. Only 22% of adults report engaging
in regular physical activity, i.e. a minimum of 30 min of moderate-to-vigorous
activity on most days of the week, the activity level most recently recommended
by the American College of Sports Medicine, the Centers for Disease Control and
Prevention, and others (123). Some 25% of adults report that they never engage
in physical activity during leisure time.
The prevalence of regular physical activity varies according to demographic
characteristics. Men are more physically active than women (16). Only 40% of
American women participate in any form of regular physical activity (17, 18). Par-
ticipation in regular activity declines with age, with women experiencing a greater
decline in older age groups than men (18). African American and Hispanic adults
P1: VEN
June 29, 2000 22:0 Annual Reviews AR103-02

BEHAVIORAL DETERMINANTS OF EXERCISE 23

are less physically active than Caucasians (17, 123). Education and income are
also both positively associated with physical activity level (6, 16, 38, 56). Marital
status is also related to physical activity level. Unmarried people are the most
active and married women the least (123). These demographic differences sug-
gest that barriers and preferences for physical activity likely vary across different
population subgroups and are factors that need to be understood in developing
programs to increase physical activity.

CHAPTER PURPOSE

In light of the well-documented health benefits of physical activity and the fact
that the majority of adult men and women are inactive, promoting regular physical
activity is a public health priority. The goal of the chapter is to review current
research findings regarding the determinants of exercise behavior and to discuss
(a) the implications of this knowledge for individual and public health recom-
mendations and (b) intervention strategies for promoting physical activity. The
discussion is predicated on the belief that physical activity is a complex, dynamic
process. During their lives, individuals typically move through various phases of
exercise participation that are determined by diverse factors (81, 84, 108). This
chapter discusses physical activity determinants in two broad categories: indi-
vidual characteristics such as motivations, skills, and other health behaviors; and
environmental characteristics such as access, cost, and social and cultural supports.
A recurrent theme in most recent discussions of physical activity is complexity,
e.g. multiple pathways to change (5); tailoring of interventions with regard to in-
dividual, environmental, and cultural characteristics; and increasing recognition
that the determinants of physical activity at initiation, maintenance, and relapse
may differ (73). An attempt is made to capture this complexity in the review.

DIFFERENTIATING BETWEEN EXERCISE ADOPTION,

MAINTENANCE, AND RELAPSE
Physical activity, like many other health behaviors (e.g. weight loss and smoking),
is often cyclical or episodic. People begin exercise programs, participate actively
for a time, and then stop, only to resume again later. In other words, each year many
people begin exercising and many people stop exercising. One study estimated
that in a 1-year period, 34% of women adopted moderate activities and about 5%
adopted vigorous activity. Within months, however, attrition rates were about 30%
and 50% for moderate and vigorous activity, respectively (107). The cyclical nature
of physical activity has led to increasing interest in examining physical activity
determinants during different phases of this cycle. Why people initiate programs of
physical activity, what determines how long a cycle lasts, and what causes people
to stop exercising are separate questions that deserve separate consideration.
P1: VEN
June 29, 2000 22:0 Annual Reviews AR103-02

24 SHERWOOD JEFFERY

PERSONAL CHARACTERISTICS ASSOCIATED

WITH PHYSICAL ACTIVITY
Motivation
Central to understanding the determinants of physical activity is the question of
why people spend leisure time doing physical activity when it could be spent in
other ways. Health, appearance, enjoyment, social interaction, stress relief, chal-
lenge, skill development, achievement, and personal satisfaction are among the
top reasons reported for engaging in regular physical activity (40, 88). Motives
differ by gender. Women are more likely to say that social factors and release of
tension are major benefits of physical activity, whereas men tend to describe the
benefits of activity in terms of fitness and health (7, 118). Motivators for participa-
tion in physical activity may also influence peoples activity choices. Frederick &
Ryan (40) found that people who participated in individual sports were more mo-
tivated by interest and enjoyment whereas those involved in fitness activities were
more motivated by physical appearance. One study reported that some people are
more motivated to exercise by the desire to avoid unpleasant aspects of a sedentary
lifestyle (i.e. they remind themselves of the negative consequences of not exercising
and that they feel worse if they do not exercise) than they are by focusing on the en-
joyable aspects of exercise (61). Relatively little information, however, is available
regarding how motivation may vary by phase of exercise, particularly what moti-
vates individuals to initiate exercise, to discontinue exercise after periods of regular
physical activity, and to reinitiate exercise following a lapse in regular activity.
Cross-sectional survey research suggests that walking is the most popular form
of exercise in the population as a whole (44.1%), followed by gardening or yard
work (29.4%), stretching exercises (25.5%), riding a bicycle (15.4%), strength
training (14.1%), stair climbing (10.8%), jogging or running (9.1%), aerobics or
aerobic dancing (7.1%), and swimming (6.5%) (123). Men more commonly report
gardening or yard work, strengthening exercises, jogging or running, and vigor-
ous or contact sports. Walking and aerobics or aerobic dancing are most popular
among women. Little information is available, however, regarding whether there
are particular types of activities or combinations of activities that are associated
with longer-term exercise maintenance. For example, is it better for individuals
to participate in one form of activity consistently or to adopt an exercise program
with an emphasis on cross training? Additionally, what type of activity is best to
recommend for those initiating an exercise program? Walking is widely promoted
as a form of physical activity for a variety of reasons, including its ease and con-
venience and its association with positive health outcomes, regardless of intensity
(30, 114). Programs often encourage walking as an activity for people who are
beginning to exercise; however, this may not be the most effective strategy for
all individuals. Information regarding whether physically inactive persons may
be more likely to begin and maintain walking for exercise than they would other
types of activity would be beneficial (114).
P1: VEN
June 29, 2000 22:0 Annual Reviews AR103-02

BEHAVIORAL DETERMINANTS OF EXERCISE 25

Self-Efficacy
Among the psychological correlates of exercise that have been examined, exercise
self-efficacy is the strongest and most consistent predictor of exercise behavior.
Self-efficacy predicts both exercise intention and several forms of exercise behav-
ior (12, 20, 29, 47, 48, 75, 77, 80, 83, 95, 96, 104). Self-efficacy is an individuals
belief in his/her capability of executing the courses of action necessary to satisfy sit-
uational demands. It is theorized to influence the activities that individuals choose
to approach, the effort expended on such activities, and the degree of persistence
demonstrated in the face of failure or aversive stimuli (4). Exercise self-efficacy is
the degree of confidence an individual has in his/her ability to be physically active
under a number of specific/different circumstances, or in other words, efficacy
to overcome barriers to exercise (29). Self-efficacy is thought to be particularly
important in the early stages of exercise (80). In the early stage of an exercise
program, exercise frequency is related to ones general beliefs regarding physical
abilities and ones confidence that continuing to exercise in the face of barriers
will pay off. Individuals with greater self-efficacy are more likely to adhere to
exercise programs with sufficient regularity to reach a point where the behavior
has become, to a certain extent, habitual.

Stage of Change
The transtheoretical model and its application to a wide range of health behaviors,
including physical activity, has received considerable recent attention (25, 102).
The transtheoretical model is an integrative model for understanding how people
progress toward adopting and maintaining health behavior change. Core features
of the model are the five stages of change and the processes of change. The trans-
theoretical model views change as a process involving progress through a series of
stages, including precontemplation, contemplation, preparation, action, and main-
tenance. Processes of change are the strategies people use to progress through
the stages (100). Research has shown that individuals can be easily staged for
exercise (64, 76), that exercise stage is associated with exercise level (78), and that
the psychological and behavioral correlates associated with stage of change for
physical activity and with transitions from one stage to another are similar to those
seen in other health behaviors (76, 79). Assessing stage of change for exercise,
thus, provides some useful descriptive information regarding physical activity.
The predictive utility of stage of change for exercise is less clear. However, initial
research examining the efficacy of stage-matched interventions suggests that this
may be a promising strategy (72).

Exercise History
Prior history of physical activity should positively influence future physical activity
behavior by promoting and shaping self-efficacy for exercise and by developing
physical activity skills. The observed relationship between exercise history and
P1: VEN
June 29, 2000 22:0 Annual Reviews AR103-02

26 SHERWOOD JEFFERY

exercise behavior varies, however, depending on how exercise history is defined.

Recent exercise history is generally predictive of future exercise behavior (27).
Childhood exercise history, however, is inconsistently related to physical activity in
adulthood (45, 48). Childhood physical activity experiences are also only modestly
predictive of adult self-efficacy and exercise behavior (45). The perception of the
exercise experience as a child may be as important as the amount of childhood
exercise. One recent study found that recalling being forced and/or encouraged to
exercise as a child was associated with lower levels of physical activity in adulthood
(120).

Body Weight
Body weight is a strong correlate of physical activity. An abundance of cross-
sectional research shows that heavier individuals are less active than lighter indi-
viduals, and prospective research indicates that changes in physical activity level
are associated with changes in body weight in the direction predicted by the energy
balance equation (23, 26, 37, 41, 44, 47, 62, 115, 124, 125). Exercise has also been
shown to improve short- and long-term weight loss in experimental studies (62).
Clearly, body weight and physical activity are inextricably linked. The extent
to which weight status is a barrier to physical activity, a consequence of physical
activity, or a motivating factor for initiating activity is unclear. Heavier individ-
uals may be more sedentary than lighter-weight individuals in part because, for
heavier people, physical activity is less pleasurable and in part because of the
embarrassment heavier people feel about being seen in public in exercise clothes
(46). However, weight status can also be a motivator for initiating exercise. One
of the most common reasons individuals give for exercising is weight control, and
dieting to control weight is positively associated with frequency of participation
in both high- and moderate-intensity physical activity (41).

Health Risk Profiles

Given that health risk behaviors tend to cluster together and that the health care bur-
den multiplies with the increasing presence of risk factors, researchers have been
interested in examining relationships between physical activity and other health
behaviors, such as smoking, alcohol intake, and diet. Examining relationships be-
tween physical activity and other health risk behaviors has important implications
for developing an understanding of the determinants of physical activity initia-
tion, maintenance, and relapse, particularly in light of recent interest in examining
whether changes in one behavioral domain may be related to changes in another
(34). Across a number of health risk behaviors, the strongest correlates of physical
activity are smoking and diet.
Although not all studies agree (53), data from cross-sectional studies indicate
that smokers are less likely to lead physically active lifestyles than nonsmokers
(9, 34, 109, 119, 121). Not only are smokers generally less likely to be physically
active, smokers are also less likely to consider making positive changes in their
P1: VEN
June 29, 2000 22:0 Annual Reviews AR103-02

BEHAVIORAL DETERMINANTS OF EXERCISE 27

exercise patterns and are less likely to initiate an exercise program (46). Cigarette
smoking may be thought of as a barrier to physical activity because it has a dis-
ruptive effect on physical activity performance; exercise is easier for nonsmokers
than for smokers. Some researchers have begun to examine physical activity as a
gateway for smoking cessation. Should smokers interested in quitting smoking be
encouraged to start an exercise program? Does engaging in regular physical activ-
ity help individuals avoid smoking and/or do changes in smoking status influence
changes in physical activity level? One recent study randomized women smokers
to either a cognitive-behavioral smoking cessation program or the same program
and a vigorous exercise program (71). Results indicated that smokers who partic-
ipated in the physical activity program were approximately twice as likely to be
abstinent from smoking from posttreatment through 3 months of follow-up.

Diet
Active adults generally have healthier diets than do sedentary adults. Given the
energy requirements of physical activity, chronic moderate-level exercise is nec-
essarily associated with increased energy intake (8). However, physically active
adults tend to eat diets lower in fat compared with their sedentary counterparts
(47, 48). In their discussion of some of the mechanisms through which eating and
exercise behavior might interact with one another to positively influence health,
King & Tribble (62) raised the issue that the impact of exercise on appetite and
eating behavior may vary according to body weight. Exercise often is associated
with a decrease in appetite (33), particularly in the short term, and some studies
have demonstrated no change in food intake with exercise in heavier individuals
(128). For those who have difficulty managing their weight, exercise may provide
a healthy behavioral alternative to overeating. Moreover, exercise also has positive
effects on mood, which may help support decreases in energy intake as well.

Stress
High levels of stress may be associated with poor health behavior patterns, in-
cluding lower levels of physical activity (2). Cross-sectional research indicates
that those who engage in higher levels of physical activity report lower levels of
perceived stress (1). Research also suggests that physical activity has a positive
impact on mood and stress level (42, 103, 116, 129). Stress levels are not static
and must be considered dynamic processes that interact with individuals cop-
ing responses to produce behavioral outcomes. Stetson et al (117) conducted a
prospective evaluation of the effects of stress on exercise adherence, which is a
good illustration of the dynamic interaction between personal, environmental, and
behavioral domains. The study examined the impact of stress on exercise behav-
ior in a sample of female exercise maintainers. Despite the fact that the majority
of women in the sample were regular exercisers, most had reported at least one
exercise relapse in the past, and on average, participants omitted nearly 1 day per
week of planned exercise during the study period. This finding was consistent with
P1: VEN
June 29, 2000 22:0 Annual Reviews AR103-02

28 SHERWOOD JEFFERY

other literature suggesting that even regular exercisers may stop and start bouts of
regular activity and that exercise adherence may vary on a day-to-day basis. Mi-
nor stress appeared to significantly disrupt exercise adherence and exercise-related
cognitions. Perceived stress had the greatest impact on exercise behavior. The
authors raised the possibility that planned exercise may itself be a minor stres-
sor during periods of ongoing stress. If this is the case, it may be misleading to
recommend exercise as a stress management tool in all cases. Missing planned
exercise sessions may result in frustrations and dissatisfaction and decreased ex-
ercise self-efficacy. Further understanding of the dynamic relationships between
stress and exercise initiation, maintenance, relapse, and resumption would be help-
ful for informing interventions and developing strategies to help individuals cope
with barriers to exercise that are associated with high stress levels, such as time
pressure.

ENVIRONMENTAL CHARACTERISTICS ASSOCIATED

WITH PHYSICAL ACTIVITY

Social Support
Social support is another robust correlate of physical activity. Individuals who
engage in regular exercise report more support for activity from people in their
home and work environments (47, 48, 61). Exercise starters are more likely to
perceive their families as being supportive of their desire to maintain good health
(46). Additionally, individuals who joined a fitness program with their spouse
had higher rates of adherence at 12 months compared with those who joined
without a spouse (126). In a comprehensive review, Carron et al (15) examined six
major sources of social influence on physical activity, including such important
others as physicians or work colleagues, family members, exercise instructors
or other in-class professionals, coexercisers, and members of exercise groups.
The authors concluded that social influence generally has a small-to-moderate
effect. Effects that were moderate to large were found for (a) family support and
attitudes about exercise, (b) task cohesion, and adherence, (c) important others
and attitudes about exercise, and (d ) family support and compliance behavior. It
should also be recognized that the relationship between physical activity and social
support is a dynamic process in which sources of social support may change over
time and through the phases of adoption and maintenance of this health behavior
(89). Research also suggests that there may be gender differences in the effect
of social influence on physical activity (56). Troped & Saunders (122) examined
gender differences in social influence on physical activity for men and women
in various stages of exercise adoption. They found that women reported greater
motivation than men to comply with most people, my regular doctor, spouse,
and parents. Gender differences in normative beliefs and social influences were
more pronounced at earlier stages of exercise adoption.
P1: VEN
June 29, 2000 22:0 Annual Reviews AR103-02

BEHAVIORAL DETERMINANTS OF EXERCISE 29

Time
Time constraints are the most frequent barriers to exercise, reported by both seden-
tary and active individuals (28, 61). It has been suggested that as people establish
a pattern of adherence to various health-promoting behavior, less-deliberate deci-
sion making about adherence occurs and behavior becomes more habitual (68).
However, scheduling efficacy remains an important and significant predictor of
adherence even among regular exercisers (29). Therefore, to maintain exercise
adherence, regular exercisers have to become adept at dealing with time as a bar-
rier. The time barrier may be a particular problem for certain population subgroups.
For example, Schmitz et al (112) reported that becoming a parent is associated
with reductions in physical activity for mothers. Time spent caring for children
may be interfering with attempts to maintain physical activity levels.

Access
Another environmental barrier that has received some attention in the determinants
literature is access to exercise facilities. One way of assessing this has been to
examine whether the distance between individuals homes and exercise facilities
is correlated with exercise behavior (111). It appears that there is a modest re-
lationship between access to facilities and physical activity. Access to exercise
facilities may be related to exercise levels for some individuals but not for others,
depending on activity preference. For those individuals who prefer exercises such
as walking or running, which can be done anywhere, access to facilities may be
less relevant. Additionally, for those who exercise with home equipment, which
could include stationary bikes, treadmills, and even exercise videos, access to fa-
cilities may also not affect exercise adherence. Regardless, the extent to which
environments are conducive to physical activity (i.e. walking/biking paths, safe
streets) likely has a strong impact on population activity levels. One recent study
examining the association between neighborhood safety and sedentary behavior
in a population-based sample found that there was a lower prevalence of physi-
cal activity among persons who perceive their neighborhoods as unsafe. Better
measurement of environmental resources for physical activity and strategies for
improving access to physical activity facilities are needed.
Another aspect of location related to exercise initiation, adherence, and relapse
concerns whether activities are group based or individual. Initial advantages of
group exercise, such as group support and structure provided by these options,
may be outweighed by the long-term costs involved in traveling to exercise sites
at specific times for physical activity involvement (54). There is some indication
that individuals, regardless of current exercise status, are more likely to report a
preference for physical activity that could be performed on ones own rather than
with others in a group or class (61). In a population-based weight-gainprevention
trial, individuals were given opportunities to participate in both home-based and
face-to-face exercise program options. One face-to-face option included a free
P1: VEN
June 29, 2000 22:0 Annual Reviews AR103-02

30 SHERWOOD JEFFERY

1-month membership to local YWCA/YMCA facilities with free childcare. Less

than 1% of participants took advantage of free access to the YWCA/YMCA com-
pared with 12% who participated in a correspondence marathon program, in
which they were asked to complete a minimum of 26 miles of walking or running
within a 4-week period on their own (113). Several studies have examined alter-
natives to supervised group exercise (3, 22, 31, 94). In the context of a weight-loss
program, Perri and colleagues (94) compared the effects of two exercise regimens,
a group-based program versus a home-based program, on exercise participation,
adherence, and fitness. Their results indicated that participants in the home-based
program demonstrated better adherence and exercise performance data, particu-
larly at the 12-month follow-up. Two recent studies compared structured exercise
programs to a lifestyle approach and reported that the lifestyle approach was as
effective as structured exercise programs in improving physical activity and health
outcomes, including body weight and cardiorespiratory fitness (3, 31). The rele-
vance of exercise location across the phases of exercise initiation, maintenance,
and relapse deserves further attention.

Attributes of Exercise Behavior

Characteristics of the exercise behavior itself (e.g. type of exercise, variety, in-
tensity, and duration) are additional important predictors of physical activity that
deserve further investigation. Although survey data show the most popular types
of activity, little information is available regarding which types of activity are
associated with the best maintenance rates. Individuals who engage in higher
doses of activity and higher intensities of activity typically have personal profiles
predictive of exercise behavior (i.e. their self-efficacy, motivations, and valuations
of exercise are strong) (65, 105). Whether the exercise behavior itself contributes
to these strong attitudes or the attitudes preceded the exercise behavior is not clear.
Variety is another attribute that could be important for promoting physical activ-
ity. Little information is available regarding whether variety in exercise behavior
matters for exercise consistency. Engaging in a wider variety of exercises could
promote adherence by reducing injury and boredom, and by increasing the adapt-
ability of being able to engage in multiple activities. Alternatively, adopting a
regular exercise regime and focusing primarily on one activity may reduce deci-
sion making and thus facilitate adherence. Individual differences in preference for
a single activity type versus variety are also likely to exist. Preference could also
vary across exercise phase (i.e. as individuals become more fit, they may be better
able to enjoy new activities).
Finally, given the public health recommendation to accumulate 30 min of phys-
ical activity on most days of the week, there has been a recent focus on the health
benefits of short bouts versus long bouts of activity. Generally, research suggests
that significant health benefits are accrued by engaging in multiple short bouts of
activity (24, 32, 51, 52, 87). It is important to note that short bouts of activity may
be easier for people to incorporate into their schedules and may be particularly
P1: VEN
June 29, 2000 22:0 Annual Reviews AR103-02

BEHAVIORAL DETERMINANTS OF EXERCISE 31

suitable for those initiating physical activity. Research suggests that multiple
short bouts of exercise have been effective at promoting short-term adherence
(51). When compared with a home-based continuous or long-bout exercise pro-
gram with a strong behavioral component, a short-bout program was comparable
with regard to exercise adherence (52). Andersen et al (3) and Dunn et al (31)
each compared a structured exercise program to a lifestyle exercise program that
emphasized the accumulation of at least 30 min of moderate-intensity physical
activity on most days of the week. Unfortunately, however, the published reports
did not provide information regarding the duration and timing of the exercise bouts
for each of the groups, so the effect of multiple versus continuous bouts of activity
could not be assessed.

Injury
Despite the numerous health benefits that exercise confers, regular physical activity
increases risk of musculoskeletal injuries. It is surprising that, with the exception of
injury risk associated with running, there has been relatively little research in this
area (66, 67, 92). Injury risk associated with running is quite high (35%65%) and
increases with running frequency and injury history (92). A recent study examined
injury rates from a variety of moderate-intensity activities and found that injury
rates associated with walking, gardening, outdoor bicycling, aerobics, and weight
training are relatively low (98). We need to understand more about frequency and
adverse effects of injury to better inform exercise initiators, consistent exercisers,
and those who are reinitiating exercise after a relapse about injury prevention. For
those who are initiating exercise programs, we need to learn more about the best
way to become more physically active while preventing injury. The low injury risk
associated with walking, in particular, appears to be another reason to endorse this
favored form of exercise (98). We also need to learn more about the best way to re-
main injury free while engaging in a consistent exercise program. Although some
research has addressed the issue of injury-prevention strategies such as stretch-
ing (49, 92), systematic examination of relationships between injury-prevention
strategies and injury outcome in representative samples of exercisers is necessary.
Possible strategies for preventing injury that successful exercise maintainers may
practice are avoiding exercise burnout and injury by not overtraining, engaging in
stretching, flexibility, and strength training, and cross training. Moreover, since
injury is a major cause of exercise relapse (110), we need to learn more about the
best strategies for avoiding reinjury when resuming exercising.

INTERACTION OF PHYSICAL ACTIVITY

DETERMINANTS
Despite the fact that numerous questions remain, there is clearly an abundance of
information available regarding multiple determinants of physical activity. An im-
portant next step in advancing our understanding of the etiology and maintenance
P1: VEN
June 29, 2000 22:0 Annual Reviews AR103-02

32 SHERWOOD JEFFERY

of physical activity patterns is to conduct studies that incorporate these multiple

determinants and reflect the complexity of predicting exercise behavior. Specif-
ically, we need to apply methodologies that can increase our understanding of
how variables from different domains interact with one another for different in-
dividuals to influence physical activity levels. For example, King et al (59) used
signal detection methodology to address this question in a study that yielded some
interesting findings. Signal detection methodology allows for an understanding of
how groups of variables may interact to influence the probability that a behavior
will occur over time. King and colleagues applied signal detection methodology
to identify the best combination of predictors of exercise adherence in a sam-
ple of sedentary adults who were participating in randomized controlled clinical
trial that included four conditions: higher-intensity group-based exercise, higher-
intensity home-based exercise, lower-intensity home-based exercise, and assess-
ment only. Their results suggested that prediction of adherence may be enhanced
by systematic evaluation of variables from multiple domains and in combination
with one another. They were able to identify subgroups of people who showed
different patterns of exercise adherence across a 2-year period, and they showed
that determinants of exercise differed depending on type of exercise program. For
example, moderately overweight individuals did less well in group-format pro-
grams and individuals who reported higher stress levels at baseline did less well
in home-based exercise formats. Application of this methodology to the study of
physical activity determinants may provide important information for matching
participants to exercise interventions and for making recommendations for those
who are interested in initiating physical activity programs. Further application of
this and related methodologies should be used to understand the determinants of
exercise behavior across the different phases of exercise initiation, maintenance,
and relapse.

Summary and Implications for Physical Activity Promotion

Recommendations
We have attempted to review the literature on key individual level and environ-
mental level determinants of physical activity. Highlights of the review include
the importance of (a) understanding and assessing different motivations for phys-
ical activity, (b) self-efficacy as a predictor of physical activity and a target for
intervention, (c) assessing readiness for physical activity change, (d ) addressing
prominent barriers to physical activity such as time and access, and (e) enhancing
social support for physical activity. We now discuss the implications of these
results for physical activity promotion and intervention. Questions addressed in-
clude how to interest people in physical activity, how to help people find the kind
of program that is right for them, and how to make the environment maximally
supportive for physical activity. Although the physical activity intervention lit-
erature has been extensively reviewed elsewhere (36, 54, 55, 60), we use previous
P1: VEN
June 29, 2000 22:0 Annual Reviews AR103-02

BEHAVIORAL DETERMINANTS OF EXERCISE 33

intervention studies to help illustrate and support our recommendations where

appropriate.

Motivation for Physical Activity

According to prevailing conceptualizations of physical activity, to become a regular
exerciser, an individual has to adopt the belief that exercise confers enough benefits
to outweigh its costs. This may be especially difficult for individuals who have been
inactive for a long time or who have had negative experiences with physical activity,
because the benefits of an active lifestyle are perceived as much less salient than the
costs of becoming more active. One approach to promoting physical activity is to
emphasize the wide range of benefits associated with a physically active lifestyle
and to tailor this message for different subgroups of the population.
For example, exercise helps people manage their weight and prevent (or slow
down) weight gain with age. Assistance with weight management/appearance can
be an important motivation for exercise initiation and maintenance. More than two
thirds of American adults are trying to lose or keep from gaining weight. The role
that exercise can play in weight management needs more emphasis. At the same
time, it is important to keep in mind that being overweight and discomfort with
ones weight status are also barriers to becoming physically active. Therefore,
physical activity promotion programs need to be modified to address the needs of
overweight individuals as well. An important aspect of physical activity promotion
for weight control is how much to recommend. Data from studies of successful
weight-loss maintainers suggest that the optimal amount of exercise for weight
control likely exceeds current public health recommendations. Indeed, it may
be close to an hour of moderate-to-vigorous activity on most days of the week
(63, 85). This higher level of activity may be perceived as intimidating and difficult
to achieve for those who have a history of sedentary behavior and are overweight.
Strategies and support for helping individuals gradually increase their physical
activity level while remaining injury free are needed. The weight management
component of physical activity also may be important in the exercise relapse
process. Decreases in physical activity are associated with increases in weight.
However, it is not clear which comes first, the exercise relapse or the weight gain.
What role does weight gain play in reinitiating exercise?
Additional important motivations for physical activity may be its positive car-
ryover effects on other aspects of health. One recent study found that smok-
ers were more likely to successfully quit smoking if they were taking part in an
exercise program (71). Exercise may also be a helpful strategy for individuals
who are attempting to manage their use of other substances as well, such as al-
cohol or high-fat foods. Given the strong connection between physical activity
and emotional well-being, more emphasis might be placed on promoting exercise
as a stress management tool (57). Although physical activity is associated with
positive well-being generally, more attention should be paid to designing physical
P1: VEN
June 29, 2000 22:0 Annual Reviews AR103-02

34 SHERWOOD JEFFERY

activity interventions to assist with mood management and stress levels for specific
populations at risk. For example, King & Brassington (57) discuss adapting phys-
ical activity programs for adult family caregivers, a population at high risk for
negative health behavior patterns and stress levels. Physical activity programs
have also been shown to be an effective strategy for alleviating depression (86).

Self-Efficacy for Physical Activity

Addressing self-efficacy should figure prominently in efforts to promote physical
activity, given its strong association with exercise behavior. Given that self-efficacy
for exercise is such a strong predictor, McAuley et al (84) suggest that we should
target exercise self-efficacy as an outcome in and of itself. More specifically, we
need to figure out the optimal way to promote self-efficacy among individuals
with a history of sedentary behavior. Courneya & McAuley (21) have written
extensively about cognitive strategies for altering self-efficacy. They recommend
targeting various incentive-laden aspects of physical activity, such as its health
benefits, to build motivation. They note that minimizing the perception of barriers
at different stages of regular physical activity involvement is important, and that de-
veloping strategies for detecting and overcoming barriers could be part of targeted
intervention for altering an individuals perceived control over, and self-efficacy
for engaging in, physical activity. Research has demonstrated that relatively short,
acute bouts of physical activity have been shown to enhance/boost self-efficacy
(82, 84). Therefore, a vital step in enhancing exercise self-efficacy may be to
actually get people to begin exercising. Once an individual begins to exercise, re-
ceiving feedback regarding exercise performance can help maintain and promote
exercise self-efficacy cognitions. Health and fitness professionals who work with
exercise initiates can play an important role in fostering strong beliefs in exercise
capabilities. Self-monitoring can be also be used as a tool for those who exercise
on their own to monitor performance and provide evidence regarding physical
activity accomplishments (84, 127). Another strategy for enhancing exercise self-
efficacy would be to incorporate more information about the cyclical nature of ex-
ercise behavior into physical activity promotion efforts. Few exercisers adhere per-
fectly to their exercise regimes. For novice and experienced exercisers, missing
planned exercise sessions can result in frustrations and dissatisfaction and de-
creased exercise self-efficacy (117). However, if individuals are taught to reframe
these experiences as normal adjustments to the ongoing challenge of staying
committed to and overcoming barriers to physical activity, self-efficacy may be
preserved.

Readiness for Physical Activity Change

The audience most difficult to reach is people who are sedentary and have no
history of positive experience with physical activity. Such individuals tend to
have elevated risk profiles across behavioral, psychological, and physiological
domains (46) and, according to the transtheoretical model, would most likely be
P1: VEN
June 29, 2000 22:0 Annual Reviews AR103-02

BEHAVIORAL DETERMINANTS OF EXERCISE 35

classified in the precontemplation stage. Programs and public health messages

need to be tailored for this group and marketed to attract them. An important
direction has been to attempt to match physical activity intervention strategies to
individuals stage of change for exercise. For example, it has been suggested that
a major limitation of physical activity programs is that they rely too heavily on the
behavioral processes of change and therefore do not reach or appeal to those in the
audience who are currently sedentary and in the precontemplation stage. Given
that individuals in the precontemplation stage, by definition, are not considering
initiating an exercise program, messages highlighting the benefits of exercising
and the costs of a sedentary lifestyle are necessary. Research examining the utility
of such strategies has shown that stage-matched interventions are helpful in pro-
moting progression through the stages of change for physical activity and increases
in physical activity levels. Additionally, instead of waiting for such individuals to
join exercise programs or initiate physical activity on their own, developing strate-
gies such as using mail-based outreach approaches may be effective for reaching
this population (13, 54, 58, 72, 73).

Barriers to Physical Activity: Time and Access

Problem-solving training for overcoming barriers to exercise should be incor-
porated into exercise intervention programs (20). Time and access are the most
commonly reported barriers to physical activity. The lifestyle approach to physical
activity is one way of addressing both time and access barriers and is a promising
strategy for attracting sedentary individuals to exercise programs. The lifestyle
approach includes promoting lower-intensity physical activities, such as walking,
that can readily fit into a daily routine and incorporating multiple bouts of physical
activity to achieve daily goals rather than concentrating an entire days commit-
ment to a single session (51, 52). The approach reduces the need for blocks of time
and special facilities. This approach may also be perceived as less intimidating to
individuals who have either no experience or unpleasant prior experiences with
physical activity. As noted previously, recent studies have reported that lifestyle
approaches to physical activity achieve results comparable to programmed exer-
cise initiatives. However, we need to learn more about what people actually do
when they are given the lifestyle message (i.e. how they accumulate the requisite
30 min per day of activity). The lifestyle approach may not be effective for some
individuals who would benefit from personalized attention and the structure im-
posed by an organized exercise program. It should be considered that the lifestyle
approach within the context of a clinical trial, which includes contact with health
professionals and/or research staff, is different from how this approach might be
carried out by individuals who are on their own. Some concern has been raised
about whether activity levels reported by people engaging in lifestyle exercise are
of sufficient magnitude and duration to warrant enthusiastic support. Whether or
not these concerns are valid, however, it would seem that the lifestyle approach
may at least be used as a bridge to help people become more physically active and
P1: VEN
June 29, 2000 22:0 Annual Reviews AR103-02

36 SHERWOOD JEFFERY

potentially more likely to engage in the type and amount of vigorous activity that
would be associated with the greatest health benefits.

Enhancing Social Support for Physical Activity

Research data clearly establish that both reports of social support and objective
indices of support, such as exercising with a partner, predict high physical activity
levels. Women, in particular, also frequently cite social interaction as a primary
motivation for wanting to exercise. Implications of these findings are that people
should be encouraged to exercise with others and that the social aspects of physical
activity should be emphasized in physical activity programs. The choice of social
support partner is a critical issue deserving further exploration. Spouses, friends,
and work colleagues are all potential exercise partners; however, who the optimal
support partner is may be different for different individuals. At least one study
suggests that it is better to have multiple sources of support.

Environmental Strategies
A supportive environment is a prerequisite to adequate physical activity levels.
People often cite access as a barrier, and both correlational and experimental data
suggest that access is important (35, 50, 106). Not all attempts to enhance physical
activity by reducing access barriers have been successful, but the preponderance
of evidence supports the idea that public and private (e.g. work sites) support for
physical activity facilities is beneficial. In addition to modifying environments to
promote physical activity, attention should be paid to modifying environments to
reduce access to sedentary behaviors (35). Better measurement of environmen-
tal resources for physical activity and strategies for improving access to physical
activity facilities are needed. In recent years, more attention has been paid to devel-
oping environmental and policy interventions to promote physical activity (106).
Prominent among the strategies suggested are developing better walking/biking
paths for the purposes of both recreation and transportation to work, making work
sites more conducive to physical activity by providing incentives and facilities to
promote exercise (e.g. subsidizing health club memberships and providing show-
ers/dressing rooms for those who commute to work by walking or biking).
Another way in which the environment has an impact on population physical
activity levels is through its influence on social norms and knowledge regarding
physical activity (106). Research examining the impact of mass media campaigns
on physical activity levels has shown that these campaigns have limited effective-
ness (74). More research is needed to know the most effective way to communicate
public health messages regarding the importance of physical activity and strategies
for incorporating exercise into ones lifestyle. Although many people know about
the importance of exercise, there may be misconceptions about exercise (i.e. you
have to be an athlete or engage in vigorous activity to achieve health benefits).
As more research is conducted on alternative ways to obtain health benefits from
physical activity, results should be communicated to the public. We need to ensure,
however, that our public health messages are based on strong empirical findings.
P1: VEN
June 29, 2000 22:0 Annual Reviews AR103-02

BEHAVIORAL DETERMINANTS OF EXERCISE 37

Without modifying the environment and cultural milieu regarding physical ac-
tivity, efforts targeted toward getting individuals to change behavior will be less
effective.

SUMMARY AND CONCLUSION

The past century has produced dramatic changes in physical activity patterns in the
United States. Machines with motors have replaced human labor in virtually every
aspect of life, so that the energy expenditure now required for daily life is a fraction
of what it was a generation or two ago. The consequences of this dramatic change
are far reaching and only now are beginning to be carefully studied. Low levels
of energy expenditure are increasingly recognized as important contributors to a
variety of chronic health problems, including diabetes, heart disease, and cancer.
Understanding of the epidemic of sedentariness and of how to reverse it is just
beginning. This chapter reviewed research conducted to date that has been aimed at
this goal. A database characterizing the population distribution of physical activity
is evolving, and consistent patterns of intrapersonal and environmental correlates of
physical activity are emerging. These associations provide a basis for developing
hypotheses about intervention approaches. Available data on intervention efficacy
are only beginning to accumulate. We have suggested that exploring ways to in-
crease motivation for physical activity, enhance self-efficacy for exercise, enhance
social and environmental supports, and tailor interventions for different subgroups
of individuals are all important for increasing population physical activity lev-
els. We highly recommend that such strategies be evaluated using strong research
designs to assess their efficacy.

ACKNOWLEDGMENTS
This work was supported by the National Institute of Diabetes and Digestive and
Kidney Diseases, grant number DK50456, and by the National Heart, Lung and
Blood Institute, grant number HL41332.

Visit the Annual Reviews home page at www.AnnualReviews.org

LITERATURE CITED

1. Aldana SG, Sutton LD, Jacobson BH, Quirk
MG. 1996. Relationships between leisure
time physical activity and perceived stress.
Percept. Mot. Skills 82(1):31521
2. Allison KR, Adlaf EM, Ialomiteanu A,
Rehm J. 1999. Predictors of health risk be-
haviours among young adults: analysis of
the National Population Health Survey. Can.
J. Public Health 90(2):8589
3. Andersen RE, Wadden TA, Bartlett SJ,
Zemel B, Verde TJ, Franckowiak SC. 1999.
Effects of lifestyle activity vs structured aer-
obic exercise in obese women: a random-
ized trial. JAMA 281(4):33540
4. Bandura A. 1986. Social Foundations of
Thought and Action: A Social Cognitive
Theory. Englewood Cliffs, NJ: Prentice Hall
5. Baranowsk T, Anderson C, Carmack C.
P1: VEN
June 29, 2000 22:0 Annual Reviews AR103-02
38 SHERWOOD JEFFERY
1998. Mediating variable framework in
physical activity interventions. How are we
doing? How might we do better? Am. J.
Prev. Med. 15(4):26697
6. Bauman A, Owen B, Rushworth RL. 1990.
Recent trends and socio-demographic de-
terminants of exercise participation in Aus-
tralia. Community Health Stud. 14(1):19
26
7. Biddle S, Bailey C. 1985. Motives toward
participation and attitudes toward physical
activity of adult participants in fitness pro-
grams. Percept. Mot. Skills 61:83134
8. Blair SN, Ellsworth NM, Haskell WL,
Stern MP, Farquhar JW, Wood PD. 1981.
Comparison of nutrient intake in middle-
aged men and women runners and controls.
Med. Sci. Sports Exerc. 13:31015
9. Blair SN, Jacobs DR, Powell KE. 1985.
Relationships between exercise or physical
activity and other health behaviors. Public
Health Rep. 100:17280
10. Blair SN, Kampert JB, Kohl HW, Barlow
CE, Macera CA, et al. 1996. Influences
of cardiorespiratory fitness and other pre-
cursors on cardiovascular disease and all-
cause mortality in men and women. JAMA
276(3):20510
11. Bouchard C, Shephard RJ, Stephens T,
eds. 1994. Physical Activity, Fitness, and
Health: International Proceedings and
Consensus Statement. Champaign, IL:
Hum. Kinet.
12. Brawley LR, Rodgers WM. 1993. Social
psychological aspects of fitness promotion.
In Exercise Psychology, ed. P Seraganian,
pp. 25498. New York: Wiley
13. Cardinal BJ, Sachs ML. 1996. Effects of
mail-mediated, stage-matched exercise be-
havior change strategies on female adults
leisure-time exercise behavior. J. Sports
Med. Phys. Fitness 36(R2):1007
14. Carpenter CL, Ross RK, Paganini-Hill
A, Bernstein L. 1999. Lifetime exercise
activity and breast cancer risk among
post-menopausal women. Br. J. Cancer
80:185258
15. Carron AV, Hausenblaus HA, Mack D.
1996. Social influence and exercise: a
meta-analysis. J. Sport Exerc. Psychol. 18:
116
16. Caspersen CJ, Merritt RK. 1992. Trends
in physical activity patterns among older
adults: the Behavioral Risk Factor Surveil-
lance System, 19861990. Med. Sci. Sports
Exerc. 24(4):526
17. Caspersen CJ, Merritt RK. 1995. Physi-
cal activity trends among 26 states, 1986
1990. Med. Sci. Sports Exerc. 27(5):713
20
18. Casperson CJ, Merritt RK, Health GW,
Yeager KK. 1990. Physical activity pat-
terns of adults aged 60 years and older.
Med. Sci. Sports Exerc. 22(Suppl.):S79
19. Costas P, Garcia-Palmieri MR, Nazario E,
Sorlie PD. 1978. Relation of lipids, weight
and physical activity to incidence of coro-
nary heart disease: the Puerto Rico Heart
Study. Am. J. Cardiol. 42(4):65358
20. Courneya KS, McAuley E. 1993. Can
short-range intentions predict physical ac-
tivity participation. Percept. Mot. Skills
77(1):11522
21. Courneya KS, McAuley E. 1995. Cognitive
mediators of the social influence-exercise
adherence relationship: a test of the the-
ory of planned behavior. J. Behav. Med.
18(5):499515
22. Craighead LW, Blum MD. 1989. Super-
vised exercise in behavioral treatment for
moderate obesity. Behav. Ther. 20:4960
23. Dannenberg AL, Keller JB, Wilson PWF,
Castelli WP. 1989. Leisure-time physi-
cal activity in the Framingham Offspring
Study: description, seasonal variation, and
risk factor correlates. Am. J. Epidemiol.
129:7688
24. DeBusk RF, Stenestrand U, Sheehan M,
Haskell WL. 1990. Training effects of long
versus short bouts of exercise in healthy
subjects. Am. J. Cardiol. 65(15):101013
25. DiClemente CC, Prochaska JO. 1982. Self-
change and therapy change of smoking
behavior: a comparison of processes of
P1: VEN
June 29, 2000 22:0 Annual Reviews AR103-02
BEHAVIORAL DETERMINANTS OF EXERCISE
change in cessation and maintenance. Ad-
dict. Behav. 7(2):13342
26. DiPietro L, Williamson DF, Caspersen CJ,
Eaker E. 1993. The descriptive epidemiol-
ogy of selected physical activities and body
weight among adults trying to lose weight:
the Behavioral Risk Factor Surveillance
System Survey, 1989. Int. J. Obes. Relat.
Metab. Disord. 17(2):6976
27. Dishman RK, Sallis JF. 1994. Determi-
nants and interventions for physical activ-
ity and exercise. See Ref. 11, pp. 214
38
28. Dishman RK, Sallis JF, Orenstein DR.
1985. The determinants of physical ac-
tivity and exercise. Public Health Rep.
100(2):15871
29. DuCharme KA, Brawley LR. 1995. Pre-
dicting the intentions and behavior of ex-
ercise initiates using two forms of self-
efficacy. J. Behav. Med. 18(5):47997
30. Duncan JJ, Gordon NF, Scott CB. 1991.
Women walking for health and fitness.
How much is enough? JAMA 266(23):
329599
31. Dunn AL, Marcus BH, Kampert JB, Garcia
ME, Kohl HW, Blair SN. 1999. Compari-
son of lifestyle and structured interventions
to increase physical activity and cardiores-
piratory fitness: a randomized trial. JAMA
281(4):32734
32. Ebisu T. 1985. Splitting the distances of
endurance training: on cardiovascular en-
durance and blood lipids. Jpn. J. Phys.
Educ. 30:3743
33. Edholm OG, Fletcher JG, Widdowson EM,
McCance RA. 1955. The energy expendi-
ture and food intake of individual men. Br.
J. Nutr. 9:286300
34. Emmons KM, Marcus BH, Linnan L, Rossi
JS, Abrams DB. 1994. Mechanisms in mul-
tiple risk factor interventions: smoking,
physical activity, and dietary fat intake
among manufacturing workers: Work-
ing Well Research Group. Prev. Med.
23(4):48189
35. Epstein LH. 1998. Integrating theoretical
39
approaches to promote physical activity.
Am. J. Prev. Med. 15(4):25765
36. Eyler AA, Brownson RC, King AC, Brown
D, Donatelle RJ, Heath G. 1997. Phys-
ical activity and women in the United
States: an overview of health benefits,
prevalence, and intervention opportunities.
Women Health 26(3):2749
37. Folsom AR, Caspersen CJ, Taylor HL, Ja-
cobs DR, Luepker RV, et al. 1985. Leisure
time physical activity and its relationship to
coronary risk factors in a population-based
sample: the Minnesota Heart Survey. Am.
J. Epidemiol. 121(4):57079
38. Folsom AR, Cook TC, Sprafka JM, Burke
GL, Norsted SW, Jacobs DR. 1991. Dif-
ferences in leisure-time physical activ-
ity levels between blacks and whites in
population-based samples: the Minnesota
Heart Survey. J. Behav. Med. 14(1):19
39. Frank CW, Weinblatt E, Shapiro S, Sager
RV. 1966. Physical inactivity as a lethal fac-
tor in myocardial infarction among men.
Circulation 34(6):102233
40. Frederick CM, Ryan RM. 1993. Differ-
ences in motivation for sport and exercise
and their relations with participation and
mental health. J. Sports Behav. 16:12446
41. French SA, Jeffery RW, Forster JL, Mc-
Govern PG, Kelder SH, Baxter J. 1994.
Predictors of weight change over two years
among a population of working adults: the
Healthy Worker Project. Int. J. Obes. Relat.
Metab. Disord. 18(3):14554
42. Gauvin L, Rejeski WJ, Norris JL. 1996.
A naturalistic study of the impact of acute
physical activity on feeling states and affect
in women. Health Psychol. 15(5):39197
43. Gauvin L, Spence JC. 1996. Physical activ-
ity and psychological well-being: knowl-
edge base, current issues, and caveats. Nutr.
Rev. 54(4, Pt. 2):S5365
44. Gibbons LW, Blair SN, Cooper KH, Smith
M. 1983. Association between coronary
heart disease risk factors and physical fit-
ness in healthy adult women. Circulation
67(5):97783
P1: VEN
June 29, 2000 22:0 Annual Reviews AR103-02
40 SHERWOOD JEFFERY
45. Hoftstetter CR, Hovell MF, Sallis JF. 1990.
Social learning correlates of exercise self-
efficacy: early experiences with physical
activity. Soc. Sci. Med. 31(10):116976
46. Hooper JM, Veneziano L. 1995. Distin-
guishing starters from nonstarters in an em-
ployee physical activity incentive program.
Health Educ. Q. 22(1):4960
47. Hovell M, Sallis J, Hofstetter R, Barring-
ton E, Hackley M, et al. 1991. Identi-
fication of correlates of physical activ-
ity among Latino adults. J. Community
Health 16(1):2336
48. Hovell MF, Sallis JF, Hofstetter CR, Spry
VM, Faucher P, Caspersen CJ. 1989. Iden-
tifying correlates of walking for exercise:
an epidemiologic prerequisite for physical
activity promotion. Prev. Med. 18(6):856
66
49. Jacobs SJ, Berson BL. 1986. Injuries to
runners: a study of entrants to a 10,000 me-
ter race. Am. J. Sports Med. 14(2):15155
50. Jakicic JM, Wing RR, Butler BA, Jeffery
RW. 1997. The relationship between pres-
ence of exercise equipment in the home and
physical activity level. Am. J. Health Pro-
mot. 11(5):36365
51. Jakicic JM, Wing RR, Butler BA, Robert-
son RJ. 1995. Prescribing exercise in mul-
tiple short bouts versus one continuous
bout: effects on adherence, cardiorespira-
tory fitness, and weight loss in overweight
women. Int. J. Obes. Relat. Metab. Disord.
19(12):893901
52. Jakicic JM, Winter C, Lang W, Wing RR.
1999. Effects of intermittent exercise and
use of home exercise equipment on ad-
herence, weight loss, and fitness in over-
weight women: a randomized trial. JAMA
282:155460
53. Johnson MF, Nichols JF, Sallis JF, Cal-
fas KJ, Hovell MF. 1998. Interrelationships
between physical activity and other health
behaviors among university women and
men. Prev. Med. 27(4):53644
54. King AC. 1994. Community and public
health approaches to the promotion of
physical activity. Med. Sci. Sports Exerc.
26(11):140512
55. King AC. 1997. Intervention strategies and
determinants of physical activity and exer-
cise behavior in adult and older adult men
and women. World Rev. Nutr. Diet. 82:148
58
56. King AC, Blair SN, Bilds DE, Dishman
RK, Dubbert PM, et al. 1992. Determi-
nants of physical activity and interventions
in adults. Med. Sci. Sports Exerc. 24(Suppl.
6):S22136
57. King AC, Brassington G. 1997. Enhanc-
ing physical and psychological function-
ing in older family caregivers: the role of
regular physical activity. Ann. Behav. Med.
19(2):91100
58. King AC, Carl F, Birkel L, Haskell WL.
1988. Increasing exercise among blue-
collar employees: the tailoring of work-
site programs to meet specific needs. Prev.
Med. 17(3):35765
59. King AC, Kiernan M, Oman RF, Kraemer
HC, Hull M, Ahn D. 1997. Can we iden-
tify who will adhere to long-term physical
activity? Signal detection methodology as
a potential aid to clinical decision making.
Health Psychol. 16(4):38089
60. King AC, Rejeski WJ, Buchner DM. 1998.
Physical activity interventions targeting
older adults: a critical review and recom-
mendations. Am. J. Prev. Med. 15(4):316
33
61. King AC, Taylor CB, Haskell WL, DeBusk
RF. 1990. Identifying strategies for increas-
ing employee physical activity levels: find-
ings from the Stanford/Lockheed Exercise
Survey. Health Educ. Q. 17(3):26985
62. King AC, Tribble DL. 1991. The role of ex-
ercise in weight regulation in nonathletes.
Sports Med. 11:33149
63. Klem ML, Wing RR, McGuire MT, Seagle
HM, Hill JO. 1997. A descriptive study of
individuals successful at long-term main-
tenance of substantial weight loss. Am. J.
Clin. Nutr. 66(2):23946
64. Laforge RG, Velicer WF, Richmond RL,
P1: VEN
June 29, 2000 22:0 Annual Reviews AR103-02
BEHAVIORAL DETERMINANTS OF EXERCISE
Owen N. 1999. Stage distributions for five
health behaviors in the United States and
Australia. Prev. Med. 28(1):6174
65. Lox CL, Burns SP, Treasure DC, Wasley
DA. 1999. Physical and psychological pre-
dictors of exercise dosage in healthy adults.
Med. Sci. Sports Exerc. 31(7):106064
66. Macera CA, Pate RR, Davis DR. 1989.
Runners health habits, 1985the Ala-
meda 7 revisited. Public Health Rep. 104:
34149
67. Macera CA, Pate RR, Powell KE, Jack-
son KL, Kendrick JS, Craven TE. 1989.
Predicting lower-extremity injuries among
habitual runners. Arch. Intern. Med. 149:
256568
68. Maddux JE. 1993. Social cognitive models
of health and exercise behavior: an intro-
duction and review of conceptual issues. J.
Appl. Sport Psychol. 5:11640
69. Magnus K, Matroos A, Strackee J. 1979.
Walking, cycling, or gardening, with or
without seasonal interruption, in relation
to acute coronary events. Am. J. Epidemiol.
110(6):72433
70. Manson JE, Hu FB, Rich-Edwards JW,
Colditz GA, Stampfer MJ, et al. 1999. A
prospective study of walking as compared
with vigorous exercise in the prevention of
coronary heart disease in women. N. Engl.
J. Med. 341(9):65058
71. Marcus BH, Albrecht AE, King TK, Parisi
AF, Pinto BM, et al. 1999. The efficacy of
exercise as an aid for smoking cessation
in women: a randomized controlled trial.
Arch. Intern. Med. 159(11):122934
72. Marcus BH, Banspack SW, Lefebvre RC,
Rossi JS, Carleton RA, Abrams DB. 1992.
Using the stages of change model to in-
crease the adoption of physical activity
among community participants. Am. J.
Health Promot. 6(6):42429
73. Marcus BH, Bock BC, Pinto BM, Forsyth
LH, Roberts MB, Traficante RM. 1998. Ef-
ficacy of an individualized motivationally-
tailored physical activity intervention. Ann.
Behav. Med. 20(3):17480
41
74. Marcus BH, Owen N, Forsyth LH, Cavill
NA, Fridinger F. 1998. Physical activity in-
terventions using mass media, print media,
and information technology. Am. J. Prev.
Med. 15(4):36278
75. Marcus BH, Pinto BM, Simkin LR,
Audrain JE, Taylor ER. 1994. Application
of theoretical models to exercise behavior
among employed women. Am. J. Health
Promot. 9(1):4955
76. Marcus BH, Rakowski W, Rossi JS. 1992.
Assessing motivational readiness and deci-
sion making for exercise. Health Psychol.
11(4):25761
77. Marcus BH, Selby VC, Niaura RS, Rossi
JS. 1992. Self-efficacy and the stages of
exercise behavior change. Res. Q. Exerc.
Sport 63(1):6066
78. Marcus BH, Simkin LR. 1993. The stages
of exercise behavior. J. Sports Med. Phys.
Fitness 33(2):8388
79. Marcus BH, Simkin LR, Rossi JS, Pinto
BM. 1996. Longitudinal shifts in employ-
ees stages and processes of exercise be-
havior change. Am. J. Health Promot.
10(3):195200
80. McAuley E. 1992. The role of efficacy cog-
nitions in the prediction of exercise behav-
ior in middle-aged adults. J. Behav. Med.
15:6588
81. McAuley E. 1993. Self-efficacy and the
maintenance of exercise participation in
older adults. J. Behav. Med. 16:10313
82. McAuley E, Courneya KS, Lettunich J.
1991. Effects of acute and long-term exer-
cise on self-efficacy responses in sedentary,
middle-aged males and females. Gerontol-
ogist 31(4):53442
83. McAuley E, Jacobson L. 1991. Self-
efficacy and exercise participation in
sedentary adult females. Am. J. Health Pro-
mot. 5(3):18591
84. McAuley E, Lox C, Duncan TE. 1993.
Long-term maintenance of exercise, self-
efficacy, and physiological change in older
adults. J. Gerontol. 48(4):21824
85. McGuire MT, Wing RR, Klem ML, Hill
P1: VEN
June 29, 2000 22:0 Annual Reviews AR103-02
42 SHERWOOD JEFFERY
JO. 1999. Behavioral strategies of indi-
viduals who have maintained long-term
weight losses. Obes. Res. 7(4):33441
86. Moore KA, Blumental JA. 1998. Exercise
training as an alternative treatment for de-
pression among older adults. Altern. Ther.
Health Med. 4(1):4856
87. Murphy MH, Hardman AE. 1998. Train-
ing effects of short and long bouts of brisk
walking in sedentary women. Med. Sci.
Sports Exerc. 30(1):15257
88. Nies MA, Vollman M, Cook T. 1998. Facil-
itators, barriers, and strategies for exercise
in European American women in the com-
munity. Public Health Nurs. 15(4):263
72
89. Oka RK, King AC, Young DR. 1995.
Sources of social support as predictors of
exercise adherence in women and men ages
50 to 65 years. Womens Health Res. Gen-
der Behav. Policy 1:16175
90. Paffenbarger RS, Hyde RT, Wing RR,
Hsiech CC. 1986. Physical activity, all-
cause mortality, and longevity of college
alumni. N. Engl. J. Med. 314(10):60513
91. Paffenbarger RS, Wing AL, Hyde RT.
1978. Physical activity as an index of heart
attack risk in college alumni. Am. J. Epi-
demiol. 108(3):16175
92. Pate RR, Macera CA. 1994. Risks of exer-
cising: musculoskeletal injuries. See Ref.
11, pp. 100818
93. Pate RR, Pratt M, Blair SN, Haskell WL,
Macera CA, et al. 1995. Physical activ-
ity and public health. A recommendation
from the Centers for Disease Control and
Prevention and the American College of
Sports Medicine. JAMA 273(5):4027
94. Perri MG, Martin AD, Leermakers EA,
Sears SF, Notelovitz M. 1997. Effects of
group- versus home-based exercise in the
treatment of obesity. J. Consult. Clin. Psy-
chol. 65(2):27885
95. Poag K, McAuley E. 1992. Goal set-
ting, self-efficacy, and exercise behavior.
J. Sport Exer. Psychol. 14:35260
96. Poag-DuCharme KA, Brawley LR. 1993.
Self-efficacy theory: use in the prediction
of exercise behavior in the community
setting. J. Appl. Sport Psychol. 5:17894
97. Powell KE, Blair SN. 1994. The public
health burdens of sedentary living habits:
theoretical but realistic estimates. Med.
Sci. Sports Exerc. 26(7):85156
98. Powell KE, Heath GW. 1998. Injury rates
from walking, gardening, weightlifting,
outdoor bicycling, and aerobics. Med. Sci.
Sports Exerc. 30(8):124649
99. Powell KE, Thompson PD, Caspersen CJ,
Kendrick JS. 1987. Physical activity and
the incidence of coronary heart disease.
Annu. Rev. Public Health 8:25387
100. Prochaska JO, DiClemente CC. 1983.
Stages and processes of self-change of
smoking: toward an integrative model of
change. J. Consult. Clin. Psychol. 51(3):
39095
101. Prochaska JO, DiClemente CC. 1984.
Self change processes, self efficacy and
decisional balance across five stages of
smoking cessation. Prog. Clin. Biol. Res.
156:13140
102. Prochaska JO, Velicer WF, Rossi JS,
Goldstein MG, Marcus BH, et al. 1994.
Stages of change and decisional balance
for 12 problem behaviors. Health Psy-
chol. 13(1):3946
103. Reed J, Berg KE, Latin RW, La Voie JP.
1998. Affective responses of physical ac-
tive and sedentary individuals during and
after moderate aerobic exercise. J. Sports
Med. Phys. Fitness 38(3):27278
104. Rodgers WM, Brawley LR. 1993. Using
both self-efficacy theory and the theory of
planned behavior to discriminate adherers
and dropouts from structured programs. J.
Appl. Sport Psychol. 5:195206
105. Rodgers WM, Gauvin L. 1998. Hetero-
geneity of incentives for physical activ-
ity and self-efficacy for highly active and
moderately active women exercisers. J.
Appl. Soc. Psychol. 28:101629
106. Sallis JF, Bauman A, Pratt M. 1998. En-
vironmental and policy interventions to
P1: VEN
June 29, 2000 22:0 Annual Reviews AR103-02
BEHAVIORAL DETERMINANTS OF EXERCISE
promote physical activity. Am. J. Prev.
Med. 15(4):37997
107. Sallis JF, Haskell WL, Fortmann SP,
Vranizan KM, Taylor CB, Solomon DS.
1986. Predictors of adoption and mainte-
nance of physical activity in a community
sample. Prev. Med. 15(4):33141
108. Sallis JF, Hovell MF. 1990. Determinants
of exercise behavior. Exerc. Sport Sci.
Rev. 18:30730
109. Sallis JF, Hovell MF, Hofstetter R. 1992.
Predictors of adoption and maintenance
of vigorous physical activity in men and
women. Prev. Med. 21:23751
110. Sallis JF, Hovell MF, Hofstetter CR, El-
der JP, Faucher P, et al. 1990. Lifetime
history of relapse from exercise. Addict.
Behav. 15(6):57379
111. Sallis JF, Hovell MF, Hofstetter CR, El-
der JP, Hackley M, et al. 1990. Distance
between homes and exercise facilities re-
lated to frequency of exercise among
San Diego residents. Public Health Rep.
105(2):17985
112. Schmitz MKH, Jacobs DR, French S,
Lewis CE, Caspersen CJ, Sternfeld B.
1999. The impact of becoming a parent
on physical activity: the CARDIA Study.
Circulation 99(8):1108 (Abstr.)
113. Sherwood NE, Morton N, Jeffery RW,
French SA, Neumark-Sztainer D, Falkner
NH. 1998. Consumer preferences in for-
mat and type of community-based weight
control programs. Am. J. Health Promot.
13(1):1218
114. Siegel PZ, Brackbill RM, Heath GW.
1995. The epidemiology of walking for
exercise: implications for promoting ac-
tivity among sedentary groups. Am. J.
Public Health 85(5):70610
115. Slattery ML, Mcdonald A, Bild DE, Caan
BJ, Hilner JE, et al. 1992. Associations
of body fat and its distribution with di-
etary intake, physical activity, alcohol,
and smoking in blacks and whites. Am.
J. Clin. Nutr. 55(5):94349
116. Steptoe A, Kimball J, Basford P. 1998.
43
Exercise and the experience and appraisal
of daily stressors: a naturalistic study. J.
Behav. Med. 21(4):36374
117. Stetson BA, Rahn JM, Dubbert PM,
Wilner BI, Mercury MG. 1997. Prospec-
tive evaluation of the effects of stress
on exercise adherence in community-
residing women. Health Psychol. 16(6):
51520
118. Summers J, Machin V, Sargent G. 1983.
Psychosocial factors related to marathon
running. J. Sports Psychol. 5:31431
119. Tang J, Muir J, Lancaster T, Jones L,
Fowler G. 1997. Health profiles of current
and former smokers and lifelong abstain-
ers. OXCHECK Study Group. J. R. Coll.
Physicians London 31:3049
120. Taylor WC, Blair SN, Cummings SS,
Wun CC, Malina RM. 1999. Childhood
and adolescent physical activity patterns
and adult physical activity. Med. Sci.
Sports Exerc. 31(1):11823
121. Thompson DH, Warburton DM. 1992.
Lifestyle differences between smokers,
ex-smokers and nonsmokers, and impli-
cations for their health. Psychol. Health
7:31121
122. Troped PJ, Saunders RP. 1998. Gender
differences in social influence on phys-
ical activity at different stages of ex-
ercise adoption. Am. J. Health Promot.
13(2):11215
123. US Dep. Health Hum. Serv. 1996. Phys-
ical Activity and Health: A Report of the
Surgeon General. Atlanta, GA: US Dep.
Health Hum. Serv., Cent. Dis. Control
Prevent., Natl. Cent. Chronic Dis. Pre-
vent. Health Promot.
124. Voorrips LE, Lemmink KA, van Heuve-
len MJ, Bult P, van Staveren WA. 1993.
The physical condition of elderly women
differing in habitual physical activity.
Med. Sci. Sports Exerc. 25(10):1152
57
125. Voorrips LE, van Staveren WA, Hautvast
JG. 1991. Are physically active elderly
women in a better nutritional condition
P1: VEN
June 29, 2000 22:0 Annual Reviews AR103-02
44 SHERWOOD JEFFERY
than their sedentary peers? Eur. J. Clin.
Nutr. 45(11):54552
126. Wallace JP, Raglin JS, Jastremksi CA.
1995. Twelve month adherence of adults
who joined a fitness program with a
spouse vs without a spouse. J. Sports
Med. Phys. Fitness 35(3):20613
127. Williams P, Lord SR. 1995. Predictors of
adherence to a structured exercise pro-
gram for older women. Psychol. Aging
10(4):61724
128. Woo R, Garrow JS, Pi-Sunyer FX. 1982.
Effect of exercise on spontaneous calo-
rie intake in obesity. Am. J. Clin. Nutr.
36:47077
129. Yeung RR. 1996. The acute effects of ex-
ercise on mood state. J. Psychosom. Res.
40:12341